Blame the voters, who probably are NOT competent.
2 Likes
Except TTU was still operating under the Air Raid offense that Mike Leach had started and Kliff Kingsbury was a disciple of. If he had gone to OU and played on a team with some defense and balance, maybe he would have won an Heisman and been drafted 1 like the player he beat out at TTU, Baker Mayfield. If had gone to any Power 5 school that did not feature the pass, he might not have even started.
There are a large number of tables in the 168 page documents. Other tables do use model 6, with personal rating. The odds ratios from my earlier posts is based on the final full regression analysis in Table B.7.2R from the final, updated rebuttal table. This regression analysis table was also included in the peer reviewed NBER study. This regression analysis table includes personal rating, as well las LDC hooks. One can compare to the previous table for “baseline” that excludes LDC to estimate relative influence on admission of personal rating for LDC vs non-LDC.
We could be here all day talking about Tx Tech QBs coached by Leach who put up insane college numbers but had little, if any, meaningful impact in the NFL. In terms of arm talent, Mahomes was different. So much so that the NFL would have locked on to him if he were at an FCS school.
Why?
The Ivy’s are an athletic conference made up of highly rejective schools who have taken on an almost ‘mythological status’ to many but they are not a public utility. They should be able to craft their admissions to their mission as they see it within the bounds of the law.
5 Likes
Harvard and the other Ivy’s have never pretended that they craft admissions solely based on academic factors. It frustrates many but the Ivy’s see their mission as advancing society rather than academic knowledge and their admissions policies are driven by their institutional goals.
3 Likes
I didn’t ask for it to be regulated?
I simply stated my opinion what I feel they “should” aim for, for the benefit of the country as a whole, as good corporate citizens.
Sorry…what? What sourcing? What messaging? There may be performative signaling here and there but as has amply and accurately been pointed out in these comments and elsewhere, the Ivy’s have actually been doing nothing of the sort. Forever.Why we don’t take them as they’ve collectively and amply demonstrated for eternity is beyond me.
Richard Sherman - Stanford Class of 2010 
1 Like
Stanford was and is in the Pac 12 conference, which is an FBS conference.
2 Likes
Well, I was extrapolating to Ivy+. And then he played for the Seahawks 2011-2017, then the 49ers before retiring.
The previous poster did write “Ivy League”, not “Ivy+”.
Of course, there are other prestige privates in the FBS: Northwestern (Big Ten), Vanderbilt (SEC), Duke (ACC), Rice (AAC), Notre Dame (independent), USC (Pac 12, moving to Big Ten).
2 Likes
I also wrote very few, not zero
1 Like
I think the bigger issue with these types of comparisons is that Fortune 500 CEOs are ridiculously extreme outliers who usually graduated decades ago. The number of Fortune 500 CEOs has no relevance to expected outcome for typical students. Instead I expect their path to CEOs involves a variety of unique traits and circumstances throughout their career, as well as a lot of luck. Some persons with those particular traits and circumstances may or may not be more likely to apply to or attend certain colleges for undergrad.
It’s possible that attending a particular school may offer some degree of advantage for becoming a Fortune 500 CEO, but a higher number of CEOs from a particular undergrad college doesn’t prove this. You’d instead need to control for a variety of other relevant factors beyond just name of undergrad school. What is more clear is that attending a particular college is by no means required. Fortune 500 CEOs attend a wide variety of undergrad colleges and college types.
1 Like
I get a strong sense of deja vu while reading this thread. Go figure.
6 Likes
Of course, but that’s just one reference metric.
The Chetty paper has 13 different metrics (in Figure 1) that would generally be defined as success (though a few overlap, so it’s more like 9-10 unique metrics). In all cases, Ivy+ are overrepresented.
Part of the paper tries to model whether that over-representation is simply because of better inputs (i.e. the 18 year olds entering Ivy+ colleges are stars), or the Ivy+ itself confers an advantage, relative to attending a pretty good public flagship (the likes of UVA, UCLA, Michigan, etc.) If I interpret the paper/its data correctly (its long, complex, and hard to digest), the Chetty paper says “yes”, attending an Ivy+ does make a difference. This is notable, because the previous highwater-mark paper studying this stuff was probably Dale & Krueger (1st or 2nd), which basically argued “no”, for most cases.
I’m not endorsing the Chetty paper’s analysis, fwiw.
That said, my own prior is, “yeah, of course going to an Ivy+ is an advantage, even for the kid who coulda gone there but went to State instead.” I’ve held that view despite Dale & Krueger’s paper(s).
2 Likes
Ah. The analysis I have been discussing is laid out in Section 5.1 of the NBER paper, and I believe takes a somewhat different modeling approach from the one you used.
The post I replied to was only discussing one metric. The other metrics include things like number of presidents, supreme court justices, senators, Rhodes scholars, Macarthur “genius” grants, … I’d make a similar comment about all of these. Comparing the number of students with an extremely rare outcome at school x vs school y tells a particular student almost nothing about whether attending that school influences his/her chances of obtaining that rare outcome. You instead need to consider a variety of relevant controls to distinguish between cause and correlation, and for the examples above often need a larger and more recent sample size.
I have not read through the full study yet, but I did see that the paper discussed in the OP mentions that their findings are consistent with Dale & Krueger, including coming to similar conclusions when using the same data sample. The difference is Dale & Krueger focused on median/average income. While the paper linked in the OP instead focuses on portion with top 1% income. They aren’t looking at typical outcomes. They are looking at outliers.
2 Likes
It’s the same author using the same data sample. They just didn’t print the full table in NBER study because it had a different focus. However, you can still review with table D4 of the paper by comparing the LDC regression coeffcients in table D4 between column 5 which does not control for personal rating and column 6, which does control for personal rating. An example is below for single legacy.
No Control for Personal Rating: (White) Legacy Boost = +2.1 (8.5x odds ratio)
With Control for Personal Rating: (White) Legacy Boost = +2.3 (10.3x odds ratio)
(Quoting Data10, but foiled somewhat by the wonky quote-ing system of CC)
Comparing the number of students with an extremely rare outcome at school x vs school y tells a particular student almost nothing about whether attending that school influences his/her chances of obtaining that rare outcome. You instead need to consider a variety of relevant controls to distinguish between cause and correlation, and for the examples above often need a larger and more recent sample size.<<
Many of the measures are of elite outcomes, probably because those are easy to count (# of presidents/senators/supreme court justices from the Ivy+ schools).
But some of the metrics, and indeed, the ones that get the most focus in the paper, are much more realistic/achievable for the average student capable of admission to an Ivy+. Earnings brackets are top 25%, top 10%, top 1%, top 0.1%. Certainly most such students would hope for top 25% at least. The paper focuses a bit more on top 1%, at which Ivy+ students are overrepresented by a factor of ~10.
Also, the paper looks at grad school outcomes, with some separation between ordinary and “elite” grad schools. Many Ivy+ undergrads will end up in grad school at some point, and probably hope for an “elite” grad school.
More broadly, a Harvard grad likely has a much higher than normal chance of being a high ranking politician, journalist, or tech icon. That said, the average Harvard grad won’t attain the highest heights. But, the average Harvard grad likely does attain higher heights than those of other schools, including the fairly good set of selective publics referenced (UMich, UCLA, UVA, etc.)
1 Like